The Use of Sodium Polyacrylate to Increase Crop Production in Dry-Land Farming Kira E. Powell Advanced Science Research, Odessa High School, Odessa, WA 99159 March 20, 2011 Abstract Sodium polyacrylate (C 3 H 3 NaO 2 ), originally developed by the Dow Chemical Company, is a polymer that is a mix of sodium acrylate and acrylic acid. Commonly found in baby diapers, it can absorb 500 times its mass in water. This polymer has potential as a soil additive. The hypothesis was if sodium polyacrylate was applied to farmable soil then the yield in the experimental section would be higher than the control section. This should be reflected by an increase in wheat crop growth and water retention. Three plots were planted for the full scale test, the Control Plot, the Experimental Plot 1 with a 2.5 % sodium polyacrylate mixture and the Experimental Plot 2 with a 5.0 % sodium polyacrylate mixture. Multiple data collections took place throughout the test including plant height, plant count, water present and yield. The Control Plot resulted in 26.8 bushels per acre, the Experimental Plot 1 with 32.8 bushels per acre and the Experimental Plot 2 with 34.1 bushels per acre. The hypothesis was accepted because both Experimental Plots had a statistically higher average plant height and yield then the Control Plot. Introduction Agriculture has been the foundation of civilization since it allowed the first people to permanently settle in Mesopotamia over six thousand years ago; today it is no different. Ninety-nine percent of all of food consumed by humans comes from cropland (Lang, 2006). But many factors contribute to the success or failure of agricultural production including drought, erosion, and average annual rainfall. Today, the United States is experiencing severe cases of drought across the country. Eighteen out of 50 states including North and South Carolina, Wisconsin and Washington, have fallen victim to either hydrological droughts (drought due to deficiency in precipitation affecting local surface or subsurface water supply) or agricultural droughts (drought involving factors vital to agriculture production such as soil water deficiency) (NDMC, 2006) (Figure 1). Although a natural phenomenon, droughts are
32
Embed
The Use of Sodium Polyacrylate to Increase Crop … Use of Sodium Polyacrylate to... · The Use of Sodium Polyacrylate to Increase Crop Production in Dry-Land Farming Kira E. Powell
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
The Use of Sodium Polyacrylate to Increase Crop Production
in Dry-Land Farming
Kira E. Powell
Advanced Science Research, Odessa High School, Odessa, WA 99159
March 20, 2011
Abstract
Sodium polyacrylate (C3H3NaO2), originally developed by the Dow Chemical Company, is a
polymer that is a mix of sodium acrylate and acrylic acid. Commonly found in baby diapers,
it can absorb 500 times its mass in water. This polymer has potential as a soil additive. The
hypothesis was if sodium polyacrylate was applied to farmable soil then the yield in the
experimental section would be higher than the control section. This should be reflected by an
increase in wheat crop growth and water retention. Three plots were planted for the full scale
test, the Control Plot, the Experimental Plot 1 with a 2.5 % sodium polyacrylate mixture and
the Experimental Plot 2 with a 5.0 % sodium polyacrylate mixture. Multiple data collections
took place throughout the test including plant height, plant count, water present and yield.
The Control Plot resulted in 26.8 bushels per acre, the Experimental Plot 1 with 32.8 bushels
per acre and the Experimental Plot 2 with 34.1 bushels per acre. The hypothesis was
accepted because both Experimental Plots had a statistically higher average plant height and
yield then the Control Plot.
Introduction
Agriculture has been the foundation of civilization since it allowed the first people
to permanently settle in Mesopotamia over six thousand years ago; today it is no
different. Ninety-nine percent of all of food consumed by humans comes from cropland
(Lang, 2006). But many factors contribute to the success or failure of agricultural
production including drought, erosion, and average annual rainfall.
Today, the United States is experiencing severe cases of drought across the
country. Eighteen out of 50 states including North and South Carolina, Wisconsin and
Washington, have fallen victim to either hydrological droughts (drought due to deficiency
in precipitation affecting local surface or subsurface water supply) or agricultural
droughts (drought involving factors vital to agriculture production such as soil water
deficiency) (NDMC, 2006) (Figure 1). Although a natural phenomenon, droughts are
2
Figure 1. This chart shows how widespread
drought is in the United States as of February 1,
2011 (NDMC, 2011).
extremely costly, causing $6-$8 billion dollars loss annually for the United States (Hayes,
2004). And, unlike other natural disasters such as hurricanes and tornados, droughts have
more longstanding affects on a greater number of people. Several elements that
contribute to drought’s devastation, is the lack of predictability, the length (droughts can
last anywhere from several months to sixty years), and the wide scale of people it affects.
For example, the Dust Bowl during the 1930’s lasted eight years, affected 260 million
acres of cropland, and displaced nearly 2.5 million people (Monatana, 2009). A drought
can directly harm farmers and ranchers, with the loss of animals and crops. The ripple
consequences extending to the average person paying more for food. During the drought
in Australia in 2008 when thousands of acres of wheat was lost, prices rose from $258 to
$367 per ton in Australia and the global price of was inflated (Smith, 2008). It has been
theorized that the likelihood of disasters like this occurring could be reduced or entirely
eliminated with soil additives designed to absorb the moisture that is available. A
chemical that has such potential is sodium polyacrylate.
Sodium polyacrylate (C3H3NaO2)n, originally developed by the Dow Chemical
Company, is a polymer that is a mix of sodium polyacrylate and acrylic acid (Figure 2).
3
Commonly found in baby diapers and household cleaners, it can absorb 500 times its
mass in water and thus is classified as a hydrogel (France, 2008). A hydrogel is a
colloidal gel in which water is the dispersion medium. Sodium polyacrylate exists in
randomly coiled chains, and there is an absence of Na+ ions (salt is removed). The
negative charges on the coils repel each other causing them to unwind. Water is then
attracted to the negative ions and attached with hydrogen bonds. This phenomenon
allows 500 times the polymers weight in pure water to be absorbed, slightly less with
impure water (France, 2008). The polymer will continue to attach to water until all
negative ions are linked to water (Figure 3). These bonds are physical, not chemical,
allowing for the process to be reversed and then repeated indefinitely. Because of this
property sodium polyacrylate could be valuable soil additive.
A factor that has to be taken into consideration for every new soil additive, like
sodium polyacrylate, is cost effectiveness. Any treatment, no matter how beneficial, must
not compromise the profit of the crop. Ideally the benefit (additional profit) of the
Figure 2. Molecular structure of sodium
polyacrylate (Kelien, 2010).
Figure 3. A model of dry coiled sodium polyacrylate (left), and
an uncoiled strand bonded with water (Richer, 2007).
4
additive, whether it is a fertilizer or a pesticide, will outweigh the cost. For example, if an
additive cost $100 to seed an entire field then the additional yield resulting from its
application to the soil would need to equal or surpass the $100 in order to be considered
cost effective. In order for the sodium polyacrylate to be a viable soil additive it has to be
cost effective. Since the sodium polyacrylate would cost approximately $0.83-$1.65 to
seed one acre, the additional profit from the yield from that respective acre would have to
exceed that (ZGEPTC, 2011). The additional yield would result from the availability of
water that the chemical absorbed and the increased water potential.
Water potential is the possible amount of water a field can hold. This is also
called field capacity. This capacity changes with soil type. There are three main types:
sandy, loam and clay and they are named for the particle that is most present in its
composition. Sand particles are the largest followed by loam then clay. The size of the
resulting spaces between the particles, called capillaries, determines the suction force
exerted on water in the soil (Figure 4). Because the retention of water through suction is
less in sandy soils, sandy soils can not retain as much water resulting in a lower field
capacity. The opposite is true in clay soils. Clay is the finest of the three particles so
when soils are made up of mainly these particles, the resulting capillaries between the
particles are smaller. The smaller the capillaries, the more suction in them, thus clay soils
Figure 4. An illustration of the effect of
capillary size on suction (Boama, 2009).
5
can retain more water and have a higher field capacity. The downside to this is that water
is hard to extract from the clay soils. With sandy soils the water cannot be retained
adequately. The third type of soil is loamy or silty soil. Silt is the particle size between
clay and sand. It allows for big enough capillaries to easily release water but small
enough capillaries to provide adequate retention for growth. The relationship between
these soil types and their resulting field capacity can be seen in the water retention curve.
The soil type present at the Control and Experimental Plots was Shano Silt Loam. This
variety of soil has a high, 29.0 cm (11.4 in), available water capacity (NRCS, 2009). It
closely follows the loam line on the water retention curve (Figure 5).
The focus of this experiment was to test the effectiveness of sodium polyacrylate
as a soil additive in farmable soil to increase crop production. The hypothesis was if
sodium polyacrylate was applied to farmable soil then the yield in the experimental
section will be higher than the control section. This should be reflected by an increase in
wheat crop growth and water retention. Water retention in soil is directly linked to crop
Figure 5. The Soil water Retention Curve. The blue line, loam is
the closest to the Shono Loam Silt present in Plots (NIVAP, 2010).
6
Figure 6. The Pre-trial testing apparatus during a test
(Powell, 2010).
production, so theoretically, if water availability in farmable soil is increased for crops,
crop production should also be increased.
Materials and Methods
Pre-trials were done in the lab to determine the application method for the full
scale tests. Soil from the field that would later be used for the full scale test was collected
for laboratory tests. A small clear plastic tube was cut and wire mesh was hot glued to the
bottom of one end to prevent anything aside from water passing through. It was held
vertical, mesh down with a clamp and ring stand over a funneled graduated cylinder
(Figure 6). With this apparatus, tests were conducted to find how to position the sodium
polyacrylate in the soil (furrow or a broadcast method). The first test consisted of 100 g
of soil and 50 g of water in order to get a control without sodium polyacrylate. The test
lasted twenty four hours. The same procedure was repeated with 10 g sodium
polyacrylate to 90 g soil, (10 % sodium polyacrylate), 5 g sodium polyacrylate to 95 g
soil, (5 % sodium polyacrylate) and 1 g sodium polyacrylate to 99 g soil, (1 % sodium
polyacrylate). Each set was also tested using the two methods: 1) the polymer was mixed
into the soil, and 2) the polymer was set in as a layer at the bottom of apparatus. These
7
Figure 7. The furrow and broadcast methods as they are used in large
scale agricultural. The picture on the left is of an example of machinery
used for furrow application and the picture on the right is of a farmer
churning up the soil after a broadcast application (Piako Tractors,
2009).
would represent furrow or broadcast application method options (Figure 7). This test was
designed purely to determine which application method would be used. The various
percentages of sodium polyacrylate were used in order to insure that whatever percentage
was used in the full scale test the method would be applicable.
Furrow application was the first method considered and consisted of the chemical
being inserted into the soil right along with the seed, theoretically keeping a supply of
moisture readily available for the seed as it matures. This method is more labor intensive
due to the fact that each field has to be separately planted with different ratios of seed to
chemical. The other option available would be to use a broadcast method. Broadcasting
differs from the furrow application as the chemical is applied to the soil before the seed.
As the ground is tilled up in preparation for seeding, the chemical is applied and then the
ground is seeded; it is mixed into the soil prior to the crop. Although less labor intensive
then the furrow method, broadcasting is very imprecise. It would be hard to regulate the
amount of chemical allocated to a certain section of soil, for example a test plot, thereby
making it hard to run tests with varying ratios of chemical to seed. Due to this and the
results from the small scale tests, the furrow method was selected for the full scale tests.
The preciseness of the application and the ability to apply several different ratios of
8
chemical outweighed any disadvantages.
Sodium polyacrylate has never before been used in large scale agriculture and
because of this fact, application rates were unknown. Similar products were researched
and the application rates were based off their recommendations for their own products.
The most closely related product, ZEBA® made by Absorbent Technologies Inc., is a
starch based superabsorbent polymer that, like sodium polyacrylate absorbs
approximately 500 times its weight in water (Absorbent Technologies, 2009). The
company suggests using a 0.68 kg - 0.91 kg (1.5-2.0 lbs) application for wheat. This
figure was the starting point for the application ratios in the sodium polyacrylate
experimental test plots. A 0.68 kg (1.5 lbs) per acre application rate was used but instead
of a 0.91 kg (2.0 lbs) rate, the amount of sodium polyacrylate was doubled for a 1.36 kg
(3.0 lbs) per acre rate. It was determined that the 0.91 kg (2.0 lbs) was too close to the
0.68kg (1.5 lbs) per acre application for testing purposes.
For the full scale tests, a field located at 47.27º N and -118.85º W was selected
(Figure 8). The field to be used was a corner section of a larger, in use wheat field. The
Figure 8. The test plot locations at 47.27º N and -118.85º W outside of
Odessa, Washington. EX2 represents Experimental Plot 2, EX1 for
Experimental Plot 1, and C denotes the Control Plot (Google, 2010).
9
ground was divided into three plots, a control and two test plots. Each was 3.66 m (12.0
ft) by 22.9 m (66.0 ft) long with 0.61 m (2.0 ft) spacing between each plot. Soil samples
were taken for nitrogen, potassium, sulfate, and phosphorus. The same test was
preformed after the completion of the project in order to determine if the sodium
polyacrylate had any effect on the soil nutrients and also to confirm the soil was adequate
for growing purposes. These samples were sent to Best-Test Analytical Services in
Moses Lake, WA. Then the area was disced using a tillage attachment on the back of a
tractor. This process allowed for the soil to be broken up and mixed, making it easier to
insert the seeds. At this time fertilizer was applied to the test plots. The three individual
fields were defined using marker flags to show the boundaries.
The Control Plot was seeded with soft white spring wheat, Triticum aestivum, of
the Louise variety (Figure 9). This variety was developed and released by Washington
State University in 2005. Its attributes include superior end-use quality and high grain
yield potential. It also has a high-temperature adult-plant resistance to local races of stripe
rust, a highly destructive leaf fungus extremely prevalent in Washington, and partial
resistance to the Hessian fly, an insect that lays its eggs on wheat plants, destroying it in
the process ( Kidwell, 2005). The field was seeded with a wheat seeder attached to the
Figure 9. The application method used for all three
seedings. In this particular picture the Control Plot is
being seeded (Powell, 2010).
10
back of a tractor. The standard amount of wheat required to seed an entire acre is 27.2 kg
(60.0 lbs). However, based on the test plot sizes, the wheat was measured for a half acre
application, or 13.6 kg (30.0 lbs). The hopper was then cleaned out using a scoop and
vacuum. For the Experimental Plot 1, wheat was mixed for a half an acre application,
13.6 kg (30.0 lbs) of wheat with 0.34 kg (0.75 lb) of sodium polyacrylate, a 1:40
chemical to wheat ratio or a 2.5 % sodium polyacrylate mixture. The chemical was
weighed out using a hand held scale and then mixed in a large flat plastic tub by hand.
Because the chemical was not evenly spread within the wheat a small amount of water
was applied via spray bottle to adhere all the chemical to the wheat (Figure 10). This
sample was inserted into the hopper and the Experimental Plot 1, was then seeded.
This process was then repeated for the Experimental Plot 2. The hopper was
cleaned again, and the sample of sodium polyacrylate wheat was mixed. Once again it
was a half acre batch. It comprised of 13.6 kg (30.0 lbs) of wheat and 0.68 kg (1.5 lbs) of
sodium polyacrylate, a 1:20 chemical to wheat ratio or a 5.0 % sodium polyacrylate
mixture. A light spray bottle mist of water had to be used to adhere the chemical to the
wheat. This batch was then planted (Experimental Plot 2) and the planting stage of the
project was complete.
Figure 10. Adhering the chemical using a
spray bottle then mixing by hand to equally
distribute the chemical (Powell, 2010).
11
In each of the three plots, three different weekly data collections took place
throughout the growing season. The first was a water content test. The soil was tested
using a garden Rapitest moisture meter. This meter had a scale of 0-4 with 0 being no
moisture and 4 being completely saturated (Figure 11). The second test was a plant count.
Using a 0.25 m2 measuring square, the apparatus was dropped at random 10 times
throughout all Plots. Each time the numbers of sprouts inside the area were counted and
recorded (Figure 11). After the plants were counted the third set of data was collected in
the same square. Five plants were chosen at random and measured for plant height
(Figure 11). This resulted in 10 plant counts and 50 plant heights for each plot per week.
Also ten wheat plants were collected from each of the plots. They were taken back to the
lab and the root system was separated, dried using an oven and then massed. This would
indicate whether the plant was making additional mass because of the possibility of
additional water.
An alternate method for water content testing was also in the last week of the test.
Three samples of soil were taken from each plot using a soil probe. These were put in
separate plastic bags, sealed, and taken immediately to the lab and massed. This was
considered wet mass. The wet mass was recorded and then the plastic bags were left open
and the soil allowed to dry out (Figure 12). After 2 weeks, the samples were massed
Figure 11. The measuring device used to define the plant count area is
the wood square (left). The Rapitest moisture meter is visible in both
pictures and the plants are measured for height (right) (Powell, 2010).
12
Figure 13. A collected tarp sample (on the back of
the combine) about to be transferred to buckets and
later to sacks.(Powell, 2010).
again and recorded as dry mass. The wet mass was divided by the dry mass to find the
percent mass loss in water.
After 136 days, the wheat was harvested for each Plot. A tarp was installed in the
hopper of the combine to catch the wheat from its respective Plot. Each tarped sample
was carefully extracted by lifting and tipping it into grain sacks (Figure 13). This process
was carried out with care, insuring no grain was lost in transition from the combine to
each sack. After each Plot was harvested, the wheat was collected and identified in sacks;
the wheat sacks were weighed to find yield for each Plot.
To calculate the yield, the acreage of the test plots had to be determined. The area
of each of the Plots was 74 m2 (792 ft
2). Since an acre is 4,047 m
2 (43,560 ft
2), the
Figure 12. The drying station in the lab.
Observe the water moisture on the sides of the
bags (Powell, 2010).
13
plots calculated to 0.018 of an acre. To complete the yield calculation, the weights of the
wheat had to be converted into bushels. A bushel is 27.3 kg (60.0 lbs) of wheat. The
masses of the individual plots were divided by the 27.3 kg to get the number of bushels.
The number of bushels in the plots over the acreage of the plots was converted using the
known bushels per acre, 27.3 kg (60 lbs), to find the bushels that would have been
recorded had the plots been full size. Besides yield, kernel counts were taken for the
yield. Random 5.0 g samples of the harvested kernels were retrieved and the individual
kernels counted out. This number was recorded and the process was repeated 10 times for
each Plot’s yield. From that data, the average kernel mass was found by dividing the 5.0
g by the number of kernels.
The local precipitation information was retrieved from weather station at the
Odessa Grange Supply in town. They use an electronic gauge to measure the rainfall and
upload their information to Weather Underground. The annual and the monthly averages
for rainfall were collected and compared to previous years in order to determine if this
could have had an effect on the experiment because of more or less available water
throughout the growing season compared to other years.
All the data was analyzed and t-tests were run to find statistical differences. After
the yield was concluded for each plot the cost effectiveness was calculated to ascertain
whether this product was economically feasible in wide scale farm production.
Results
A soil test was run before the experiment to analyze the chemicals in the soil. The
results showed nitrogen levels at 17 ppm, sulfate at 5 ppm, the phosphorous levels at 30
ppm, and the potassium levels at 708 ppm in the top 0.3 m of soil (Appendix A). Soil
14
tests were run at the conclusion of the test as well. For the Control Plot the nitrogen level
was 3 ppm, the phosphorous at 32 ppm and the potassium at 939 ppm in the top 0.3 m
soil. Sodium was present at 0.12 meq/100g (Appendix B). For the Experimental Plot 1,
the nitrogen level was 4 ppm, the phosphorous level at 36 ppm and the potassium level at
795 ppm. Sodium was present at 0.12 meq/100g (Appendix C). For the Experimental Plot
2, the nitrogen level was 3 ppm, the phosphorous at 31 ppm and the potassium at 844
ppm. Sodium was present at 0.08 meq/100g (Appendix D).
The average wheat height within the Control Plot was 76.0 cm (±4.4) with a high
of 85.0 cm and a low of 65.3 cm. The average wheat height within the Experimental Plot
1 was 81.6 cm (±3.2) with a high of 83.3 cm and a low of 16.8 cm. The average wheat
height within the Experimental Plot 2 was 80.3 cm (±3.7) with the highest individual
sample 90.0 cm and lowest sample 72.0 cm (Table 1). A two tailed t-test was used to
statistically compare the plots. The difference between the average wheat height in
Experimental Plot 1 and the Control Plot was significantly different at the 99.9 %
confidence level (t = ±7.258; df = 98; p < .001). The difference between the average
wheat height in Experimental Plot 2 and Control Plot was significantly different at the
99.9 % confidence level (t = ±5.243; df = 98; p < .001). There was no significant
difference between the average wheat height in Experimental Plot 1 compared to
Table 1. Average Plant Height in the Control Plot, the Experimental Plot 1 and
Experimental Plot 2.
Plot N Ave. Height
(cm)
SD
(cm)
Var.
(cm)
Control Plot 50 76.0 ±4.42 19.5
Experimental Plot 1 50 81.6 ±3.22 10.4
Experimental Plot 2 50 80.3 ±3.77 14.2
15
Experimental Plot 2.
Within the Control Plot the average plant count was 35.0 (±6.6) with the high of
44 and a low of 24. Within the Experimental Plot 1 the average plant count was 41.7
(±2.6) with a high of 46 and a low of 38. Within the Experimental Plot 2, the average
plant count was 38.7 (±9.4) with the high of 57 and a low of 27 (Table 2). A two-tailed t-
test was used to statistically compare the plots. The difference between the average plant
count at harvest in the Control Plot and Experimental Plot 1 was significantly different at
the 95% confidence level (t = ±2.99; df = 18; p<.05). There was no significant difference
between the average plant count in Control Plot and Experimental Plot 2 or between the
average plant count in Experimental Plot 1 and Experimental Plot 2.
Root systems were taken from 10 samples from each plot. The average mass for
the Control Plot was 0.30 g (±.14) with the high of 0.53 g and the low of 0.08 g. The
average mass for the Experimental Plot 1 was 0.56 g (±0.20) with the high of 0.95 g and
the low of 0.28 g. The average root mass for the Experimental Plot 2 was 0.40 g (±.08)
with the high of 0.52 g and the low of 0.29 g (Table 3). A two-tailed t-test was used to
statistically compare the plots. The difference between the average root mass in the
Control Plot and Experimental Plot 1 was significantly different at the 99 % confidence
level (t = ±3.24; df = 18; p < .01).There was no significant difference between the
Table 2. Average Plant Count in the Control Plot, Experimental Plot 1, and Experimental
Plot 2.
Plot N Ave. Count SD Var.
Control Plot 10 35.0 ±6.60 43.56
Experimental Plot 1 10 41.7 ±2.58 6.66
Experimental Plot 2 10 38.7 ±9.41 88.55
16
average root mass in the Control Plot and Experimental Plot 2. The difference between
the average root mass in the Experimental Plot 1 and the Experimental Plot 2 was
significantly different at the 95 % confidence level (t = ±2.33; df = 18; p < .05).
Three soil samples were collected from each of the Plots at a depth of 0.35m (1ft).
They were sealed and taken back to the lab. They were dry and wet massed and the
percent water mass loss was recorded. The average water mass percent lost for the
Control Plot was 1.018 % (±.0004) with the high of 1.066 % and the low of 0.987 %. The
average percent water mass lost for the Experimental Plot 1 was 1.041 % (±.0041) with
the high of 1.508 % and the low of 0.749 %. The average percent water mass lost for the
Experimental Plot 2 was 0.726 % (±.0044) with the high of 1.216 % and the low of
0.341 % (Table 4). A two-tailed t-test was used to statistically compare the plots. There
was no significant difference between the average percent water mass lost in any of the
Plots.
Table 3. Average Root Mass in the Control Plot, Experimental Plot 1 and
Experimental Plot 2.
Plot N Ave. Mass
(g)
SD
(g)
Var.
(g)
Control Plot 10 0.30 ±0.14 0.0196
Experimental Plot 1 10 0.56 ±0.20 0.0400
Experimental Plot 2 10 0.40 ±0.08 0.0059
Table 4. Average Percent Water Mass Lost Between the Control Plot, Experimental
Plot 1 and Experimental Plot 2.
Plot N Ave. % Water
Mass Lost
SD
(%)
Var.
(%)
Control Plot 3 1.018 ±.0004 1.6×10-7
Experimental Plot 1 3 1.041 ±.0041 1.7×10-5
Experimental Plot 2 3 0.726 ±.0044 1.9×10-5
17
Kernel counts were taken for 5 g samples of the test plots. The Control average
was 166.8 (±8.7) kernels with the high of 180.0 and the low of 153.0. The Experimental
Plot 1 kernel count average was 158.2 (±8.0) kernels with the high being 174.0 and the
low of 148.0 The Experimental Plot 2 kernel count average was 143.0 (±6.2) kernels with
the high of 152.0 and the low of 133.0 (Table 5). A two-tailed t-test was used to
statistically compare the plots. The difference between the average kernel count in the
Control Plot and Experimental Plot 1 was significantly different at the 95 % confidence
level (t = ±2.301; df = 18; p < .05). The difference between the average kernel count in
the Control Plot and Experimental Plot 2 was significantly different at the 99.9 %
confidence level (t = ±7.064; df = 18; p < .001). The difference between the average
kernel count in Experimental Plot 1 and Experimental Plot 2 was significantly different at